One of the mechanisms responsible for the therapeutic failure of alkylating agents is the DNA repair protein, O6-alkylguanine-DNA alkyltransferase (AGT), encoded by the MGMT gene. AGT removes the alkyl group from 06-alkylguanine in a fast and single step reaction, thereby preventing the formation of DNA cross-links by chloroethylating agents such as BCNU. Currently, a strategy involving the inactivation of AGT by 06-benzylguanine (BG) followed by BCNU treatment has shown evidence of significantly increased antitumor effect of BCNU. Ongoing clinical trials are evaluating its efficacy in tumor chemotherapy. However, repeated administration of BG and BCNU will raise the possibility that BG resistant cells develop, subsequently, resulting in the failure of chemotherapy. In our recent studies, we selected two MMR deficient colon cancer cells for resistance to BG and BCNU and found two different mutations at amino acid 165 of AGT, to form K165E and K165N mutant AGT in these two cell lines. The cells harboring the K165 mutations have dramatically decreased AGT activity but remarkably increased resistance to BG+BCNU. Thus, we hypothesize that MMR deficiency leads to a high mutation frequency in DNA repair gene such as AGT gene and that two K165 mutant AGTs predominantly confer acquired resistance either to the combination BG+BCNU and BG+TMZ or alkylating agents alone. To test this hypothesis, it is necessary to distinguish acquired resistance caused by mutated AG from other resistance factors. This is of concern because the two BG-resistant AGTs were identified in cell lines with MMR defects. Once cells lose MMR, their sensitivity to various chemotherapeutic agents is decreased directly by impairing the ability to recognize or process DNA damage and indirectly by increasing the mutation rate throughout the genome. Therefore, it is possible that not only does mutation in AGT confer drug resistance, but other mechanisms of drug resistance as well. Thus, these specific objectives are proposed: to define whether K165 mutant AGTs are the major factor of acquired resistance to BG and BCNU, despite low AGT activity; to define whether colon cancer tumors with MMR deficiency are more likely to acquire resistance to BG+BCNU through mutations in MGMT than MMR wt tumors; and to determine whether BG-resistant AGT could be selected in the xenograft setting after mice carrying the tumor received multiple treatments with BG and BCNU. The long-term goal is to define the conditions in which MGMT mutations are observed in human tumors after clinical use of BG and BCNU. Overall, this project promises to provide novel information on the induction of BG-resistant AGT in drug treated MMR defective tumors and the impact of the altered AGT-resistance to BG+BCNU.